- Question
- Does adding rosiglitazone to metformin or a sulfonylurea increase cardiovascular risk compared with the combination of metformin plus sulfonylurea in people with type 2 diabetes?
- Design
- Multicentre, randomised, open-label, active-controlled trial enrolling 4447 patients. Mean follow-up was 5.5 years and analysis was by intention to treat, with a hazard ratio non-inferiority margin of 1.20.
- Population
- Adults with type 2 diabetes on metformin or sulfonylurea monotherapy, with a mean HbA1c of 7.9%.
- Intervention
- Rosiglitazone (n=2220) added to background metformin or sulfonylurea, started at 4 mg once daily and increased up to a maximum of 8 mg per day as needed to reach the HbA1c target.
- Control
- Combination of metformin and sulfonylurea (active control group, n=2227), with the added drug titrated up to the maximum permitted dose.
- Primary outcome
- Composite of cardiovascular hospitalisation or cardiovascular death.
- Results
- The primary endpoint occurred in 321 patients on rosiglitazone and 323 in the control group (HR 0.99, 95% CI 0.85-1.16), meeting non-inferiority. The HR was 0.84 (0.59-1.18) for cardiovascular death, 1.14 (0.80-1.63) for myocardial infarction and 0.72 (0.49-1.06) for stroke.
- Safety
- Heart failure causing hospital admission or death occurred in 61 patients on rosiglitazone and 29 on control (HR 2.10, 95% CI 1.35-3.27). Upper and distal lower limb fractures increased, mainly in women assigned to rosiglitazone.
- Conclusion
- Rosiglitazone increased the risk of heart failure and of some fractures without increasing overall cardiovascular morbidity or mortality versus standard glucose-lowering drugs, with inconclusive data on myocardial infarction.
- Why it matters
- It provided long-term cardiovascular safety evidence for thiazolidinediones after prior controversy, and reinforced the need to monitor for heart failure and fracture risk with this drug.
Card generated automatically with AI from the abstract of the original publication and the trial registry; every figure was checked against those sources. It may contain errors: verify the data at the source before using it for clinical decisions.
Generated on October 18, 2026